SUMMARY Somatic mutations in the core components of the pre-mRNA splicing complex, the spliceosome, are the most common genetic lesions in patients with myelodsyplastic syndromes (MDS). Specifically, recurrent missense mutations in the SF3B1 gene are present in 10-20% of all MDS cases. Inhibitors of SF3B1 have been developed and represent a promising new frontier for MDS therapy. We propose to develop pre-clinical models of Sf3b1 mutation, to test a novel SF3B1 inhibitor in these models, alone or in combination with azacitidine; and to perform correlative studies on a Phase I/II clinical trial of an SF3B1 inhibitor in collaboration with H3 Biomedicine. H3 Biomedicine has generated a compound, H3B 8800, that shows promising mutant- selective activity in cell lines and that is poised for clinical trials. The Ebert laboratory has developed a conditional knock-in mouse model that expresses the Sf3b1 K700E mutation, the most common mutation in MDS patients. Since SF3B1 mutations commonly co-occur with mutations in DNMT3A in MDS, we will create a model with both conditional Sf3b1 knock-in mutation and conditional Dnmt3a inactivation. We will test H3B 8800 in this model and will test the combination of H3B 8800 with azacitidine, a drug with efficacy in MDS and in TET2-mutated cases in particular. We will identify mechanisms of resistance to H3B 8800 using a genome- wide CRISPR-Cas9 screen. Finally, we will examine the safety and efficacy of H3B 8800 in MDS patients in a phase I/II clinical trial. This first-in-class agent has the potential for major clinical impact in a large fraction of MDS cases with aberrant spliceosome function due to somatic mutations. The studies described in this proposal will define the activity of the drug in different genetic backgrounds, examine the activity of the drug in hematopoietic stem and progenitor cells, identify mechanisms of therapy resistance as well as insights into the mode of action of the drug, and examine therapeutic efficacy in patients.